Author Of 2 Presentations
PS05.02 - Validation of three Secondary Progressive Multiple Sclerosis classification methods in five registries within the SPMS Research Collaboration Network
Abstract
Background
Assigning Secondary Progressive Multiple Sclerosis (SPMS) course consistently is challenging as it is based on a gradual worsening in neurological disability independent of relapses. Clinical SPMS assignment may therefore vary between registries depending on clinical practice. Consequently, a comparison of SPMS between registries would benefit from an objective definition of SPMS.
Objectives
To validate three different methods for classifying patients into Relapsing Remitting Multiple Sclerosis (RRMS) or SPMS, compared to the clinical assignment, in five European Multiple Sclerosis (MS) registries.
Methods
Data from MS registries in Czech Republic (11,336 patients), Denmark (10,255 patients), Germany (23,185 patients), Sweden (11,247 patients), and the United Kingdom (UK) (5,086 patients) were used. Patients with either RRMS or SPMS, age ≥ 18 years at index date (date with the latest Expanded Disability Status Scale (EDSS) observation) were included. Index period was 01/2017 - 12/2019. Three EDSS centric classification methods were applied; method 1: a modified real world EXPAND criteria (Kappos, L. et al., 2018. The Lancet 391(10127), 2018), method 2: the data-derived definition from Melbourne University but without pyramidal Functional Score (Lorscheider, J. et al., 2016. Brain 139(9)), method 3: the decision tree classifier from Karolinska Institutet (Ramanujam, R. et al., 2020. medRxiv, 2020.07.09.20149674). The classifications were compared to the clinical assignment, where sensitivity (SPMS as true positive), specificity (RRMS as true negative) and accuracy were calculated as similarity measurements.
Results
The overall classification performance (sensitivity, specificity, accuracy) among classifiable patients were; method 1: (0.47, 0.85, 0.79), method 2: (0.77, 0.87, 0.85), method 3: (0.84, 0.83, 0.84). The proportions of unclassifiable patients with each method were; method 1: 20.0%, method 2: 32.2%, method 3: 0%. Methods 2 & 3 provided a high sensitivity, specificity and accuracy, while method 1 provided high specificity but low sensitivity. Method 3 was the only method having no unclassifiable patients.
Conclusions
Our findings suggest that these methods can be used to objectively assign SPMS with a fairly high performance in different registries. The method of choice depends on the research question and to what degree unclassifiable patients are tolerable.
PS05.04 - Ongoing disease modifying treatment associated with mis-classification of secondary progressive as relapsing-remitting multiple sclerosis
Abstract
Background
Until recently, disease modifying treatment options for MS patients with a secondary progressive course (SPMS) were limited, leading to the common practice of off-label treatment with drugs approved for relapsing-remitting MS. We previously showed that applying objective algorithms tend to increase the proportion of SPMS in MS registries, suggesting that SPMS is under-diagnosed in clinical practice, possibly related to available treatment options.
Objectives
To compare characteristics of patients clinically assigned an RRMS course that are re-classified when an algorithm-based SPMS assignment method is applied.
Methods
Data from MS registries in the Czech Republic (11,336 patients), Denmark (10,255 patients), Germany (23,185 patients), Sweden (11,247 patients) and the United Kingdom (5,086 patients) were used. Inclusion criteria were patients with relapsing remitting (RR)MS or SPMS with age ≥ 18 years at the beginning of the study period (1 January 2017 – 31 December 2019). In addition to clinically assigned SPMS a data-driven assignment method was applied in the form of a decision tree classifier based on age and last EDSS (Ramanujam, R. et al., 2020. medRxiv, 2020.07.09.20149674).
Results
Across the five registries 8,372 RRMS patients were re-assigned as SPMS (Denmark: n=1,566, Czech Republic: n=1,958, Germany: n=2,906, Sweden: n=648, United Kingdom: n=1,294) increasing the overall SPMS proportion from 17% to 31%. Re-assigned patients tended be younger, were older at onset and had experienced a quicker progression to SPMS. The overall proportion of clinically assigned SPMS patients on disease modifying treatments (DMTs) was 36% but varied greatly between registries (Czech Republic: 18%, Denmark: 35%, Germany: 50%, Sweden: 40%, and the United Kingdom: 12%) whereas a higher proportion of 69% (OR=4.0, P<0.00004) were on DMTs among RRMS patients re-assigned as SPMS (Czech Republic: 71%, Denmark: 68%, Germany: 78%, Sweden: 80%, and the United Kingdom 40%).
Conclusions
SPMS patients on DMTs may be clinically mis-classified as RRMS, most likely by not being re-assigned to SPMS after conversion has occurred. This challenges the use of time to SPMS conversion as an outcome in comparative effectiveness studies using real world evidence data and argues for the use of objective classification tools in the analysis of MS patient populations.
Author Of 2 Presentations
P0009 - ENTIMOS: a discrete event simulation model for maximizing efficiency of infusion suites in centres treating multiple sclerosis patients (ID 1630)
Abstract
Background
Multiple sclerosis patients are treated with intravenous (IV) disease modifying treatments (DMTs). Infusion suite resources are thus vital components of MS patient care. Infusion suites may be dedicated to MS patients, or shared with patients with other neurological conditions, or other patients requiring infusion. Here, we describe a resource utilization model for the infusion suites of Charing Cross (UK), which serves patients with different neurological conditions.
Objectives
To maximize the clinical efficiency of the infusion suite based on three resource constraints: percentage of patients IV MS DMTs, number of infusion chairs, and number of nurses. Efficiency gains in the infusion suite may benefit both patients and the healthcare system.
Methods
ENTIMOS, a discrete event simulation (DES) model, was created using SIMUL8 based on qualitative information from infusion centers and populated with data specific to the Charing Cross hospital neurology infusion suite. Posology, administration information, and rates of immune related-reactions (IRRs) were applied from published data sources from both MS and non-MS DMTs of interest.
The infusion suite model assumes 75 MS and 21 non-MS patients weekly, including up to seven MS patients initiating IV treatment; is equipped with 12 infusion chairs and six beds; and is staffed with a total of six nurses. We simulated the effects of changing the three resource constraints described on the number of patients waiting for an appointment (queue size), the time for patients to get an appointment for their first or subsequent IV treatments (waiting times), and general resource utilization.
Results
Changing the number of chairs, moving a subset of patients from IV to any non-IV alternative treatments, moving patients between MS IV treatments, or changing the allocation of nurse resources may all have an impact on the queue size and waiting times. Once the changes are implemented in the model, existing resources optimised and the queue size reduced, the effective centre throughput can be increased.
Conclusions
ENTIMOS allows users to optimize their use of constrained resources in an infusion suite to improve patient experience and infusion suite efficiency.
P0482 - Objective classification methods result in an increased proportion of secondary progressive multiple sclerosis in five patient registries (ID 1120)
Abstract
Background
Secondary progressive MS (SPMS) is a research area that is attracting more attention as better treatment options are still needed for this patient group. The assignment of SPMS by clinicians can differ between countries and may be influenced by drug prescription guidelines, reimbursement issues and other societal limitations.
Objectives
To compare the clinically assigned SPMS proportion to three objective SPMS classification methods in five MS registries.
Methods
Data from MS registries in the Czech Republic (CR) (11,336 patients), Denmark (10,255 patients), Germany (23,185 patients), Sweden (11,247 patients) and the United Kingdom (UK) (5,086 patients) were used. Inclusion criteria were patients with relapsing remitting (RR)MS or SPMS with age ≥ 18 years at the beginning of the index period (1 January 2017 – 31 December 2019). In addition to clinically assigned SPMS three different classification methods were applied; method 1: modified real world EXPAND criteria (Kappos et al, Lancet 2018:391; 1263-1273), method 2: the data-derived definition from Melbourne University without the pyramidal Functional Systems Score (Lorscheider et al, Brain 2016:139; 2395-2405) and method 3: the decision tree classifier from Karolinska Institutet (Ramanujam, R. et al., 2020. medRxiv, 2020.07.09.20149674).
Results
The SPMS proportions per registry, when comparing the clinically assigned SPMS with the results of the three classification methods, were CR: 8.8%, 21.3%, 22.1%, 25.0%; Denmark: 15.5%, 27.5%, 25.4%, 28.0%; Germany: 15.6%, 15.4%, 16.7%, 25.4%; Sweden: 23.7%, 20.8%, 23.2%, 24.6% and UK: 34.3%, 21.7%, 38.4%, 58.3% for clinical SPMS and methods 1, 2 and 3, respectively.
Conclusions
The proportion of clinically assigned SPMS patients varies between MS registries. When applying other classification methods, the SPMS proportion generally increases but remains variable between registries. As some of the classification methods have extensive requirements regarding data density, the number of unclassifiable samples created are considerable for some of the registries, which will influence the results. Providing a classification method that depends on objective information could prove useful when attempting to estimate the proportion of SPMS patients in MS populations but the choice of method may depend on the data characteristics of the individual MS registry.